The blind source separation based on the compressed sensing
Yang Bo, Lijun Liu · 2012
For the problem of blind source separation (BSS) with the sparsity properties of the high frequency wavelet transform coefficients, the paper proposed a new method based on compressed sensing (CS) and K-means clustering algorithm. Compared with the traditional methods of blind source separation, simulation results demonstrated that the proposed method improves the quality of the recovered signal significantly, and improves the speed of separating and reconstruction obviously.